Tell us your FTP 'differential' (indoor to outdoor)

As per the thread title. What was the ‘difference’ that you found you had to make (adjusting up or down) between measured outdoor and indoor FTP. This speaks to very many threads on this forum where a large proportion of people have found their indoor value needed adjusting down to make it effective for TR training zones.

Add your numbers and % change to the list below so we can get a good archive of the various measured ‘differentials’. This should not only be interesting to see per-se’ but also provide a useful gauge for people who are starting out anew, as to what adjustments the ‘average’ user needed to make.

Username / Outdoor / Indoor / % change

dsirrom / 332w / 300w / -10%

I’m about to do my second indoor Ramp test having just finished the first part of Short Power Build so (hopefully) there will be a small increase :smiley:


1 Like

How are you measuring your outdoor ftp?

1 Like

I don’t have a measured outdoor FTP since that would require a different protocol than a ramp test, but I do notice that it is easier for me to generate power on an indoor trainer.

Outdoors - Power meter (left sided crank) - 20 minute test on a slightly inclined road that gets steeper (not by much) towards the end.

Indoors - ramp test on Wattbike Atom in ERG mode


indoor: no power meeter, Tacx Bushido Smart, ERG mode = 243 FTP
outdoor: Quark dZero, 20min - 5% = 285 FTP


I’ve never done an outside FTP test and have no place where I can get 20 mins of uninterrupted road and grade.


Jschief - Outdoor: 370-375 Stages power meter Indoor: 382 is the most I’ve done for 20 minutes. Outdoor FTP is probably a little low due to me not having a nice climb or other appropriate piece of road to perform a proper test. Granted, a nice fan makes a big difference for me when training indoors. Assuming the temperature was identical, I would like to think that outdoors I could do around 380ish for 45 minutes to an hour

Differential for me is usually about 3% lower indoors than outdoors , same powermeter.

1 Like

Same power meter, (and this is an estimate, I haven’t tested outdoors in years) but I’m about 5-7% higher outside I think. This is just based on threshold workouts that I’ve done outside lately. I think I’m about 300w +/- outdoors and my indoor FTP seems pretty close at 286w.


I’ve considered testing outdoors as I do outdoor workouts enough, but I think I can gauge pretty closely with those longer 10 min threshold intervals. I should re-test soon.

-Indoor 222
-Outdoor 233
5% difference

Same power meter

I realize this thread like many is just interesting or anecdotal but, I don’t see how comparing in and out has any usefulness to those new to power, training or cycling when there is no mention of controlling protocol or accuracy of power. Not using the same protocol with the same pm yields two unrelated numbers.

Sorry for being a wet blanket. I’m posting for the new guy who reads a thread a day here about FTP and confused as mountain biker racing a criterium.


FTP is FTP. Fix your cooling, and they’ll be the same…

That’s a big one, but not the only potential difference. Inertia (flywheel size & speed) and motivation can be large factors too.

1 Like

Sorry. Nope.


Thanks @Landis - I started the thread as I thought it would be interesting - it got relatively little response overall which perhaps answers my question :grinning:

I think the points you make may be entirely correct, (they certainly sound correct) so not a wet blanket at all. :+1: :+1:

My thinking and reason for starting the thread was that my perception of the reality on a practical level was I had noticed in many threads (and from my own experience), commentary that made me believe that a large proportion of TR users who have previously measured their FTP outdoors (regardless of method or equipment) experienced a negative differential when they begin indoor training and therefore needed to reduce the number. (notwithstanding a few to the contrary).

I completely accept that the FTP / power your body is capable of making ought to be constant, if everything else is equal, but I also acknowledge that in many cases everything else (cooling and all the myriad of other variables that people commonly cite) is unfortunately not equal between indoors and outdoors and for that reason there is (anecdotally at least) a reasonably common difference across platforms / power meter types / trainers / setups that seems (at least to me) to be more than incidental.

There is even a thread (I can’t find it to cite it right now so apologise) that seems to recognise this as a ‘thing’ and commiserates people.

So … I just thought it may be useful to document the ‘average’ differential to get some handle on what an average quantum of ‘drop’ might be (again probably due to differences in the various variables, some of which you have referenced) - I thought that might be helpful for the next guy or gal who comes along, having done an outdoor power meter text and got an X# FTP as they could then look at this information and seek to make an initial adjustment based on that average to see if it helped.

I didin’t know anything about indoor training when I started may TR journey and fell into the trap of assuming that the FTP number I would use indoors would be the same as that which I measured outdoors - I was wrong and would have saved some time if I could have accessed this type of information beforehand, just to get a rough steer.

I know that this is what the ‘Ramp’ test is supposed to be for, to give you a new baseline set of numbers for indoor training that are specific and unique for you and your own indoor setup, however I found the ramp test skewed things up for me (I have crap endurance so my (relative) strength in short power gave me a reading that was too high. I know this is just another set of variables to consider and my tiny brain can’t figure out the interrelationship between all of this stuff so apologies if I am so wide of the mark that I need a map to get back to the correct answer :rofl:

The cooling thing is certainly a big factor and I did notice a small improvement when I added two massive fans instead of the small one I had been using, but it didn’t make up all the difference so I guess that in my own circumstance there will be other things at play, such as difference in PM calibration and possibly loads of other small seemingly incidental things that add up to make a material difference.

Anyway, that was me just trying to explain my thinking in starting this thread - I guess from reading some of the points here that this is probably flawed thinking due to lack of knowledge on my part so fair enough - I will get my coat :smiley: :coat:

1 Like

Any reason you didn’t do both tests with the Quarq meter?

I don’t have a fixed differential. It depends on the length of the test. Over a full hour they would be similar, but over a short ramp test they would be huge.

If you specified 20 minutes unramped I reckon it’d be 10W, but don’t know anywhere I could cycle for 20 minutes continuous near me so have never tested it.

Edited for autocorrect oddities

Trying to make sense of it all is a good thing. We are after the same thing so, just to be clear, I’m not disagreeing trying to figure out a differential is a bad thing. As long as it’s reasonably accurate. Reasonably accurate because what gets lost is that physiology and bio mechanics are two different things. Whether on a trainer or outside the whole point is to physiologically spend time at certain intensities to elicit adaptive response by the body.

The question in my mind then becomes how do we measure this and unless I’m just in the weeds here, heart rate can be a good proxy to correlate power from multiple test methods and PM’s. Not talking about the HR you might see at the end of a ramp test vs. what you would average during a 20 minute protocol. That, for example, while doing 2x20’s on various trainers, or various bikes, HR will be constant while power will vary. Bio mechanically, you could have an FTP of 350 and 320 yet, physiologically your HR at hose wattages is the same.

It’s not that neat and tidy of coarse due to all the things that affect HR. But the concept of two vastly different power numbers can give the same physiological response is what’s important. And ultimately how you verify those numbers are doing what we want them to do physiologically?

1 Like

I have dedicated bike for trainer and one for outdoor! Bikes are pretty similar and fitted the same (with Retul)!

I would be happy to have the same numbers so I can use outdoor trainings but also the analyze…